| > consider putting some money on whoever they aren't voting for I probably should, honestly - at least if most/all of those people are voting the same way in an election. Since it's not prescience, I suspect I just don't have a good sense of the dropout rate; I learned this because 2016 was an upset, so several Clinton voters mentioned that even a seemingly-safe election had maintained their losing streak. > I imagine the rate is considerably lower for real people, who by definition are more likely to vote for the winner. This I'm not sure about, and it's a pretty interesting question. You're obviously right that a randomly-chosen person has a >50% chance of having voted for the winner, but I think this is a non-standard population issue. People have a lot of different voting strategies, and presumably the people I know are some sort of 'perverse voters' - whatever forces lead >50% of people to vote for a given candidate tend to push them in the opposite direction. I'm not sure how common this is, but I expect it's probably bigger than the random rate. If somebody's "vote for the underdog" instinct hits 60%, that would ~triple the odds of getting this outcome. > they vote for third party candidates I didn't count anyone who consistently votes third party, since that's effectively "didn't vote" for odds of winning. But I think several of these people did vote Nader in 2000, which dodged a particularly hard-to-call election. |